Weighted Frequent Subgraph Mining Using Affinity Weighing in Distributed Environment

Graph based data representations are getting popular in areas like bioinformatics, social networks, web data mining, etc. Extracting frequent subgraphs from a huge set of graphs is a fundamental task in numerous information mining applications. In areas such as mobile communication networks, social networks, etc. weighted graphs are more useful. More relevant and specific subgraphs are generated through weighted frequent subgraph mining. Most of the current techniques are memory-based and are not scalable. This work uses an existing distributed approach for Frequent Subgraph Mining using iterative MapReduce based framework and applies Affinity Weighing (AW) scheme to it, and compares it with Average Total Weighing (ATW) scheme.